Goals-of-care discussion in older adults: a clinical and ethical approach

Abstract

Opening Vignette Mdm Wong, a 75-year-old Chinese woman, was premorbidly wheelchair bound and required moderate assistance in activities of daily living. She had a history of advanced Alzheimer’s dementia, unrevascularised triple vessel disease and advanced chronic kidney disease (CKD), and was admitted for severe pneumonia with fluid overload and oliguric acute-on-CKD. Despite an initial trial of medical therapy, she continued to deteriorate clinically. In consultation with cardiology and nephrology specialists, the patient was deemed unsuitable for invasive coronary intervention or initiation of dialysis. You were the medical officer in her general medicine team, and was tasked by your ward consultant to broach goals-of-care discussion with her family as she lacked mental capacity for decision-making and had no former documentation of her expressed treatment preferences. Given the patient’s clinical and premorbid status, the medical recommendation was that she would not be a suitable candidate for extraordinary life-sustaining interventions, such as cardiopulmonary resuscitation, intubation/mechanical ventilation or intensive care unit admission, for inotropic support. However, during the discussion, the patient’s only son was insistent on “doing everything necessary” to keep his mother alive and was angry at the suggestion that the medical team might choose not to resuscitate his mother.WHAT IS GOALS-OF-CARE DISCUSSION? Goals-of-care (GOC) discussion in healthcare settings refers to a series of conversations held with patients and/or their families to explore patients’ goals, values and preferences in end-of-life treatment/care and, in the process, help to consolidate their understanding of their current state of health and prepare them on what to expect in the disease trajectory.[1] The GOC discussion is a fairly broad term that may include establishing resuscitation status in inpatient settings, advance care planning (ACP) and advance medical directives. While GOC can be discussed in younger persons with chronic debilitating or life-limiting conditions or as part of general ACP to prepare for unforeseen life events, for the purposes of this article, we will focus on the approach to GOC discussions for older adults who are aged ≥60 years according to the United Nations definition of older persons.[2] DEFINITION OF KEY TERMS Resuscitation (‘code’) status refers to clinical documentation of the extent of resuscitation to be attempted in a hospitalised patient in the event of cardiopulmonary arrest, which usually comprises the following three components:[3] (1) Code status — (a) full code: all resuscitative measures based on advanced cardiac life support will be carried out; (b) partial/limited code: patient-specific resuscitative interventions may be withheld (e.g., cardiopulmonary resuscitation (CPR), defibrillation, inotropic support, intubation/mechanical ventilation) and/or ‘maximum’ siting of care (e.g. general ward, high-dependency or intensive care unit [ICU]); and (c) do not resuscitate (DNR): no resuscitative measures will be attempted; (2) Rationale for code status — patient’s clinical condition (i.e. serious, life threatening, debilitating), comorbidities (including life-limiting illnesses, end-stage organ diseases, etc.), patient preferences (either verbally expressed during present admission or from previous ACP documentations); (3) Whether the code status was discussed with the patient and/or the family members. Advanced care planning refers to a non-legally binding, longitudinal process of exploring the goals, values and preferences in adult persons of any age and health status to guide future decisions on health and personal care in the event of a serious illness; individuals may nominate a healthcare spokesperson/surrogate decision-maker to make healthcare-related decisions for the patient in future if he/she no longer has the mental capacity to do so.[4,5] Advanced medical directives refers to a legal document completed by a mentally competent adult aged ≥21 years to state that he/she is not for extraordinary life-sustaining treatment (LST) in the event of terminal illness, where such interventions may merely delay death with no reasonable chance of meaningful recovery.[5,6] Lasting power of attorney (LPA) refers to a legally binding document in which a donor (person making the LPA) formally appoints a donee (surrogate decision-maker) to make decisions pertaining to personal welfare, including health and personal care, and/or financial/property matters, in the event the donor becomes mentally incapacitated.[5] The four pillars of medical ethics are as follows:[7] (1) Autonomy (‘Respect for Persons’): refers to a moral obligation to respect the rights of individuals with mental capacity to make healthcare-related decisions for themselves (e.g. obtaining informed consent before proceeding with medical treatments/procedures and maintaining patient confidentiality); (2) and (3) Beneficence and non-maleficence: Beneficence refers to the obligation of healthcare professionals to act in the ‘best interests’ of the patients, and non-maleficence (primum non nocere) requires the physician to do no harm to the patients; as most treatment options have inherent risks of complications or adverse effects, it is important to weigh the benefits and risks of healthcare interventions in the context of the individual patient before making recommendations that will likely provide net benefit and minimal harm; (4) Justice: refers to fairness in resource distribution (distributive justice), respect for individual rights that prohibit discrimination (rights-based justice) and respect for the rule of law (legal justice). HOW IS THIS RELEVANT/IMPORTANT TO MY CLINICAL PRACTICE? The ageing population has led to increased burden of chronic diseases, cognitive impairment and disability. It is known that older adults with poor premorbid condition and function generally do not have good resuscitation outcomes. For example, inpatient CPR involving elderly patients has poor survival outcomes, with reportedly 17% surviving till discharge.[8] These outcomes further worsen with increasing age, presence of medical comorbidities such as heart failure (HF), chronic obstructive pulmonary disease (COPD), CKD, liver cirrhosis, neurological conditions and malignancy, as well as poor premorbid functional status [Table S1, Supplemental Digital Appendix at https://links.lww.com/SGMJ/A110]. Moreover, there are also significant physical (e.g. CPR-related bony fractures or visceral injuries, defibrillation-related burns and ventilator-associated complications) and psychological complications (e.g. mental health sequelae in post-ICU syndrome) that can result from the resuscitation process and subsequent ICU admission [Table S2, Supplemental Digital Appendix at https://links.lww.com/SGMJ/A110]. As such, it is clearly prudent for clinicians to identify patients who would benefit from a conservative approach to GOC, such as those with advanced age, poor medical comorbidities and functional status, and hold timely GOC discussions with them. However, such discussions in hospital settings have often been either inadequate or suboptimally carried out. A US study reported that 66% of seriously ill patients had no code status discussions, and conversations on LSTs were often brief, lacking information on prognosis and resuscitation outcomes, offered no medical recommendations and frequently utilised medical jargons without proper explanations.[9] Locally, a study published in 2011 initially reported a lack of end-of-life discussions in the general wards,[10] which led to the implementation of a ‘physician order form’ that significantly increased DNR orders with improved engagement of patients and families in end-of-life discussions pertaining to LSTs.[11] Interestingly, in both studies, there is an observation that end-of-life discussions were commonly held with families instead of patients themselves, which was a phenomenon likely reflective of the family-oriented approach to healthcare decision-making in Asian populations.[11] However, the study cautioned that decision-making by family surrogates carries the risks of misinterpretation of a patient’s actual goals/preferences and inaccurate understanding of a patient’s clinical prognosis, as well as places undue emotional burden on family members who are making important health decisions on behalf of a loved one.[11] CLINICAL CONSIDERATIONS While the concept of ACP is generally beneficial for all persons regardless of age and health status, we highlight two main groups of patients who would particularly benefit from a timely GOC discussion, to avoid having to eventually undergo medically inappropriate/futile and potentially harmful treatment interventions at the end of life. Firstly, patients who are at high risk of clinical deterioration should have an early discussion on the appropriate GOC, given the time-sensitive nature. An assessment of patients’ risk of clinical deterioration can be made by reviewing their premorbid conditions, functional status as well as their current clinical condition based on the severity of the presenting complaint, vital signs and parameters, presence of end-organ dysfunction and treatment response.[12] Secondly, there are patients in whom aggressive LSTs are unlikely to be beneficial or in line with their GOC and may, in fact, cause harm or prolong suffering for the patients and their families. This typically includes patients of advanced age, with serious medical comorbidities and poor functional status[13] that will lower their chances of meaningful recovery from cardiopulmonary arrest. In general, a ‘serious illness’ may be defined as a health condition with high mortality risk, which either adversely affects the patient’s quality of life or functional status or leads to excessive caregiver strain.[14] Common triggers that prompt a GOC discussion in these patients may include a negative response to the ‘surprise’ question (i.e. ‘would the physician be surprised if the patient were to die within 1 year?’), observed disease progression, decline in clinical/functional status and repeated/severe hospitalisations.[15] ETHICAL CONSIDERATIONS The four governing principles of medical ethics include justice, beneficence, autonomy and non-maleficence.[7] All four ethical concepts are relevant in GOC discussion, where clinicians should respect the treatment/care preferences expressed by a patient with mental capacity, seek to recommend treatments that are in patients’ best interests and not cause harm/prolonged suffering, as well as be fair and just in resource allocation to end-of-life patients. In the context of GOC discussion, ethical dilemmas often arise when there are disputes/conflicts between patients/families and the healthcare team in terms of treatment decisions/extent of care, or when decisions need to be made for mentally incapacitated patients.[16] Firstly, it is not uncommon to have patients/families insist on all forms of aggressive treatments even if they are considered to be medically futile.[16,17,18] In brief, medical futility can be defined as treatment that is highly unlikely to lead to a beneficial outcome which the patient will have the capacity to appreciate.[17] This may refer to LSTs that merely serve to preserve/prolong the physiological bodily functions of an otherwise unconscious patient or those that will not realistically lead to continued survival beyond the acute care setting of the hospital.[17] In such cases, it is important to recognise that while autonomy means that patients have the right to refuse treatment/investigations (hence the need for informed consent), this does not extend to the right to demand treatments that are medically inappropriate/futile.[19] In addition, for patients with evidence/concern of cognitive impairment, an assessment needs to be made of their mental capacity to make healthcare decisions, based on four core tenets: understanding (comprehension and retention of information), appreciation (grasping the implication of the decision to their lives), reasoning (weighing the presented choices) and communication of decision.[20] For patients who do not have mental capacity, it is important to assess if there is a nominated surrogate decision-maker, for instance, based on previous ACP, to hold healthcare discussions with.[21] Ultimately, when faced with ethical dilemmas pertaining to end-of-life treatment decisions and GOC, which may be made more difficult when patients no longer have the mental capacity, clinicians have the obligation to provide treatment and care that are in the patients’ best interests (principles of beneficence and non-maleficence). To this end, the Jonsen’s four-box ethical model[22] is an excellent toolkit that can be used to guide clinical decision-making in such circumstances. The four key components of the Jonsen’s four-box toolkit include the following: Medical indications: to consider patient’s current medical issues and whether they are acute/chronic, reversible/irreversible, emergent/non-urgent, assess patient’s clinical prognosis and explore possible treatment options and alternatives, with evaluation of likelihood of success in the patient’s context (e.g. age, premorbid conditions and functional status) Quality of life: to clarify patient’s premorbid function, lifestyle and activities and assess the expected functional recovery/quality-of-life outcome measures with or without treatment/intervention Patient preferences: to assess patient’s mental capacity and legal competency to make healthcare decisions and ensure that adequate information on the treatment benefits, risks and alternatives has been disclosed to the patient; in cases where the patient has no mental capacity, evaluate if there is a nominated surrogate decision-maker (e.g. in previous ACP) and/or whether there are previously documented/known patient preferences, values and goals with regards to treatment/care Contextual features: to consider if there are familial issues, third-party interests/conflicts of interests and undue influence in the decision-making process and assess if there are psychosocial (e.g. religious beliefs, traditions/cultural norms, financial concerns) and/or legal issues affecting decisions For complex, ethically challenging cases that are difficult for clinicians to handle, referrals to the hospital clinical ethics committee should be considered.[16] A FIVE-STEP PRACTICAL APPROACH We herein propose a simple five-step ‘SPEAR’ model [Figure 1 and Table 1], developed with reference to the local ACP framework,[21] SPIKES template for breaking bad news,[23] Jonsen’s four-box ethical toolkit[22] and published guides on end-of-life discussions for physicians.[15,24] We further share practical tips on implementation, communication strategies and common pitfalls to avoid when using this approach for GOC discussion [Table S3, Supplemental Digital Appendix at https://links.lww.com/SGMJ/A110].Figure 1: Outline of the ‘SPEAR’ model for goals-of-care discussion.Table 1: The five-step ‘SPEAR’ model for goals-of-care discussion.When there are conflicts in terms of extent of care between patients/families and the healthcare team, clinical and ethical considerations should be reviewed. However, it is important to recognise that when patients/families unequivocally demand for ‘everything to be done’, this often represents an underlying emotional need. For instance, terminally ill patients may fear abandonment, such that they perceive foregoing LSTs as losing access to medical care,[17,18] and patients’ families may also perceive foregoing LSTs as being at odds with their moral obligation of filial piety.[18,25] Therefore, clinicians should attempt to elicit and acknowledge the underlying emotional concerns and provide the necessary reassurance and psychosocial support. BARRIERS/CHALLENGES AND PROPOSED SOLUTIONS There are various barriers to GOC discussions, which may be classified into patient/caregiver-, clinician- and disease-related factors [Table S3, Supplemental Digital Appendix at https://links.lww.com/SGMJ/A110]. Firstly, patient-/caregiver-related barriers/challenges may include unreadiness to discuss end-of-life matters,[26] difficulty accepting news of poor prognosis,[27] demands for medically ‘futile’ treatment interventions,[18] inability to comprehend the implications and limitations of LSTs,[27] disagreements within the family on GOC[27] and patient’s lack of mental capacity to make healthcare decisions.[27] Uniquely, in Asian cultures, discussion of death/dying is often considered a taboo subject that should be avoided, and filial piety is an important Confucian value that obligates children to respect and care for their elders.[28] As such, familial decision-making is often prioritised over individual autonomy, where children may collude to hide unfavourable diagnoses from their elderly parents[28] and feel obligated to pursue LST to save their parents’ lives at all cost.[25] Secondly, clinician-related barriers may be related to lack of confidence, personal discomfort or lack of time to hold GOC discussions.[26] Finally, the key disease-related hindrance to GOC discussion is the difficulty with accurate disease prognostication.[15] For instance, unlike advanced cancers that demonstrate steady clinical decline over weeks to months, chronic diseases like HF or COPD are often characterised by a fluctuating and more unpredictable clinical course that leads to difficult prognostication.[29] There are practical strategies that may be helpful to mitigate these aforementioned barriers to GOC discussion [Table S4, Supplemental Digital Appendix at https://links.lww.com/SGMJ/A110]. To overcome patients’/families’ reluctance to discuss end-of-life issues or difficulty in accepting poor prognosis, clinicians should aim to explore GOC early and opportunistically in suitable patients, rather than wait until the patient becomes terminally ill,[15] and in addition, apply established effective communication techniques during GOC discussions, such as the SPIKES model for breaking bad news,[23] serious illness conversations[24] and Jackson et al.’s[30] communication guide on cultivating prognostic awareness. For example, for patients who are clinically deteriorating but remain ambivalent/resistant to hearing prognostic information, Jackson et al.[30] propose an approach known as ‘naming the dilemma’, where the physician articulates both the patient’s fear of receiving/discussing bad news (demonstrating empathy) and the clinical concerns with the avoidance of such a discussion (acknowledging the clinical urgency in cultivating clinical awareness). In addition, methods to tackle clinicians’ lack of preparedness to initiate GOC conversations include practical simulation training[31] and use of clinical guides.[24] CONCLUSION In summary, GOC discussions form a crucial part of clinical practice in an ageing patient population with increasing frailty, disability and medical comorbidities. A simple five-step ‘SPEAR’ model can be useful for broaching GOC conversations, bearing in mind that there can be clinical barriers and ethical issues to navigate during such discussions. TAKE-HOME MESSAGES Clinical discussions on the extent of care are increasingly important in an ageing population with increasing life expectancy, but with higher burden of medical comorbidities, frailty and disability. A simple approach to GOC discussion using the ‘SPEAR’ model includes patient selection, exploring patient perspectives and treatment/care preferences, patient on disease and resuscitation outcomes, acknowledge the expressed and demonstrate and making a medical recommendation on code status and future treatment There are clinical barriers and factors and disease-related to GOC discussion that need to be An important governing ethical for GOC discussions would be respect for patient’s autonomy to make decisions and treatment/care this does not extend to the right to demand medically treatment dilemmas can be to the hospital clinical ethics committee and by using the Jonsen’s four-box Vignette You the to your as that the patient was medically not a suitable candidate for but her son with and there was risk of between the medical team and the patient’s As such, a family was for a discussion between the patient’s son and the team, including the the the patient’s son his understanding that his mother was ill and she had previously expressed not to undergo for her kidney However, that would be for not being filial to his mother if everything that was medically possible was not consultant and his expressed and that given patient’s poor premorbid conditions and function, LSTs would most likely merely prolong the process and suffering, with likelihood of further complications from resuscitation, instead of any meaningful clinical/functional Therefore, and suffering would be and care for the The patient’s son and the medical team of their continued to provide the best care for the the patient continued to treatment for her terminal and in the hospital in the presence of her support and of There are no conflicts of for

Topics & Concepts